52 REPORT — 1901. 



to the general rule. After allowing for the supports, a small difference still 

 remained between the instruments, the mean apparent amplitude of dis- 

 turbed mo%'ements being some 10 per cent, greater for No. 31 than for No. 9. 

 During the comparison the observer, Mr. Constable, noticed that on 

 certain days of high wind the trace from the seismograph standing on the 

 floor showed numerous small movements, many possessing distinct 

 asymmetry. Further investigation showed that these undoubtedly arose 

 from vibrations set up in the building by tlie gusts of wind. Minuter 

 examination showed that the phenomenon also occurred, though to a 

 much smader extent, in the traces from the seismograph on the cement 

 bed. Wind is thus clearly a cause of not infrequent tiny movements, 

 whose source had hitherto escaped detection. 



IV. Movements of Horizontal Pendulums in relation to Barometric Pressure. 



For many years it has been recognised that there is a relationship 

 between the movements of horizontal pendulums and fluctuations in 

 barometric pressure.^ 



An important and apparently practical addition to our knowledge on 

 this subject has recently been made by Mr. F. Napier Denison, of 

 Victoria, B.C., in a contribution to the Royal Meteorological Society, 

 entitled 'The Seismograph as a Sensitive Barometer.' The instrument 

 referred to is the one adopted by the British Association. Briefly stated, 

 Mr. Denison's conclusion is that the pendulum swings towards the area 

 of greatest barometric pressure. For example, it has been found that 

 when a storm area is approaching from the westward the boom of the 

 pendulum moves steadily to the eastward, and this often occurs eighteen 

 to twenty-four hours before the local barometer begins to fall. On the 

 contrary, should there be an important high area to the West, the 

 pendulum will swing in that direction before it is possible to ascertain 

 the position of such an area on the current weather charts. 



As partial confirmation of Mr. Denison's observation, it may be men- 

 tioned that a gradual but decided movement of the Shide pendulum 

 towards the West precedes stormy weather, whilst in the Report for 1895 

 referred to above there are tables showing a close relationship between 

 displacements of pendulums in Tokio and the barometric gradients at that 

 place. 



V. An Attempt to Detect and Measure any Relative Movement of tlie 

 Ujnvay, that may noio be taking place at the Ridgeway Fault, near 

 Strata Dorsetshire. Second Report by Horace Darwin, June 1901. 



Many of the early readings have been found to be of no value, because 

 water had got into the vessels containing the oil and had blocked its free 

 passage through the pipe ; this difficulty has, we hope, been overcome by 

 making the covei's of the vessels more completely watertight. 



' See Reports on ' Earthquake and Volcanic Phenomena,' issued by the British 

 Associarion in 1883, 1885, 1887, 1888, 1892, 1893, 1895, 1896. 



For a theoretical discussion of this subject see ' Applications o£ Physics and 

 Mathematics to Seismology,' by Dr. C. Chree, Phil. Mag., March 1897, p. 185. 



